The NBA, like every sports league in the world, has been accused of being rigged multiple times in its history.
Over a decade ago, the NBA was dealing with Tim Donaghy, a referee allegedly bet on games he officiated and made calls that affected the point spread. In recent history, Ayesha Curry, wife of Golden State Warriors superstar Steph Curry, tweeted about her husband getting fouled out and ejected during the Playoffs. That Tweet would soon become prime meme material:
Nearly three years after that infamous tweet, James Harden and the Houston Rockets were critical of NBA officials, calling them out for supposed bad calls in the Playoffs against the Golden State Warriors last year and this year.
Soon after Game One of this year's Western Conference Semifinal, the NBA found that the referees for that game missed a few calls (but not all of them that Harden claims), and released a Last Two Minute Report for the public — fans, players, anyone — to review for themselves. These reports are supposed to allow for transparency on officiating; referees are human, after all, and they make mistakes. It also offers justification for the calls that went right, and generally keeps a record of every foul call and non-foul call in tightly contested games.
This is why these Last Two Minute Reports are important; one wrong call can cause a butterfly effect that changes the course of NBA history.
Which brings us to a huge question: if the referees were perfect, would those close games be turned into the Rockets' favor? Would we be talking about Harden vs. Kahwi or Lillard vs. Kahwi in the Finals, rather than another year of the Golden State Warriors running wild?
With this data publicly available, we can not only see what calls were blown in crucial games, but also quantify the referees and players that play a huge role, for better or for worse, when it matters most. In other words, are the refs really against James Harden? Or is he the one actually getting away with fouling opponents? And did any of those blown foul calls, or all of them in these reports, radically shift what could have been in the 2018-19 season?
Using data as curated by our new collaborative KG Base, we’re diving deep into the regular season post-All Star Break Last Two Minute Reports and considering what could have been in the NBA if everything were perfect… Or if that all those blown calls really mattered at all.
Scroll to the bottom for the KGBase dataset and how you can contribute to our project.
Zarba and Kirkland made a wrong call during a pivotal moment of a regular season game. And in the case of Zarba — who will be calling his sixth NBA Finals over the next week — that may come back to haunt the Warriors or Raptors.
First, who are the referees that are being reviewed in these reports? Among the outliers we reported in a preview to this feature, there is also Zach Zarba, Courtney Kirkland, and Josh Tiven, the three referees who were assigned to Game 1 of this year's Warriors and Rockets series.
Among the referees with the most incorrect non-calls — or not blowing the whistle when there should have been a foul called — Kirkland is ranked among the middle of the pack in post-All Star regular season games with 1.29 INC's per game. This means that, when he oversaw games as part of a team of referees, there was at least one bad non-foul call in every game, with a second one in every fourth game.
Ed Malloy, who is one of the refs that will work the NBA Finals this year, is notably at the top of the list for incorrect non-calls per game, averaging 2.5 INCs during the regular season.
While Kirkland ranks comfortably among the extensive list of referees, Zarba is right in the bottom-25, and Tiven ranked second to last. That means they fall under the refs who recognize fouls more than incorrectly "letting the players play," according to the NBA's self-analysis.
Kirkland also holds the (dis) honor of being a part of officiating teams that had at least one incorrect call post All-Star break. During one of his seven games, he and the rest of the officiating team made an incorrect call as determined by the league.
That one game was a tight contest between the Playoff-bound Philadelphia 76ers and the lottery-bound Chicago Bulls, and the incorrect call was made when Mike Scott hit Robin Lopez on his way up for a layup.
If Robin Lopez never gets to the line due to that blown call, never hits that one free throw, and all other points remain constant, the game goes into overtime and the Bulls may not have won. That means the 76ers could have ended the season with 53 wins instead of 52...
Zabra also was part of a game with an incorrect call through his nine games officiated. That one game was also a Bulls contest; specifically, it was the Bulls against the Atlanta Hawks in a quadruple overtime slugfest. That incorrect call came with a little over 16 seconds left to play in the fourth overtime... But at that point, the game was practically over, and we should give Zarba and company a bit of leeway given they've been working for over 3 hours and 30 minutes.
(That's longer than the length of the movie "Titanic," for context)
So, what does this data tell us? That these referees were less likely to make incorrect non-calls — meaning that they, in theory, made foul calls when they happened in crucial moments — but also, for Zarba and Kirkland, made a wrong call during a pivotal moment of a regular season game.
And in the case of Zarba — who will be calling his sixth NBA Finals over the next week — that may come back to haunt the Warriors or Raptors.
Now that we have the faces behind the fouls, what fouls are they calling or missing?
It isn't sexy, but personal fouls, offensive fouls, and shooting fouls are the most common in these Last Two Minute Reports.
Above, you see every single call and non-call in the Last Two Minute Reports after the All-Star Break. That's 3,208 calls, with the majority of them — 2,218 — being correct non-calls. Why? Because in auditing itself, the NBA is looking at every single conceivable point in the last two minutes when a foul can be called, especially when it comes to personal fouls.
That brings us back to the personal foul, the most common foul type in this sport. When defending a shot, it's pretty difficult to avoid making contact with the opposing player. And, in the last two minutes of a game, personal contact needs to be made intentionally to prolong the game.
While personal fouls are the most common foul type looked at in these last Two Minute Reports, the most common incorrect call is for offensive fouls. These are personal fouls committed by the player with the ball, such as when a player charges through a set defender for a shot.
As seen above, offensive fouls are difficult to get right, according to these reports. In post-All Star Break regular season games, there were double the incorrect offensive calls than correct ones.
Travelling, a call where possession changes without a foul being assessed to a player, has a unique foul analysis profile: after all personal foul types, travelling is the most common non-call and foul call. What's more interesting, though, is the distribution. In the Last Two Minute Reports we studied, there were 7 correct calls, 14 incorrect calls, 31 correct non-calls, and 13 incorrect non-calls.
In other words, travelling has become the "is it a catch?" rule of the NBA; human beings are still split on what constitutes as travelling, especially when fancier dribbling moves, flashy plays, and eurostep layups are more common than ever before.
And, surprisingly, everyone was well behaved; less than 1% of all fouls called were some version of Flagrants or Technical fouls.
We know the referees, we know the types of fouls being called... But what about the players?
As we said in our prelude, James Harden was fuming in this year's playoffs about how the officials called his game. "I just want a fair chance," Harden said in the postgame press conference after Game 1 against the Warriors. "Call the game the way it's supposed to be called and that's it; and I'll live with the results."
These are the same refs that saw him play all throughout the regular season, and he also appears on the Last Two Minute Reports as both a player who gets fouled... And one who commits them often.
First, Harden as the victim, or in other words, the "Disadvantaged Player." In late game situations when the score is close, a winning team gives the ball to their superstar. That superstar usually has to go to the foul line often, as the opposing team wants that player to take free throws, an unguaranteed two points, instead of an easy two-pointer or a three-point dagger.
This is called "Hack-a-Shaq", named after star center and notoriously bad free throw shooter Shaquille O'Neil. How bad was he? Well...
So bad that he became meme material. Before that signature thumbs up, Coach Greg Popovich instructed one of his players to foul Shaq mere seconds into a game. Although this was a joke (Never change, Pop), it is a good example for showing why teams would intentionally foul a player.
In late game situations, Harden was among the elite players who teams tried to pester. They fouled him often, as seen in the chart below, which lead to correct calls (he got to the free throw line, in most cases) or incorrect calls (he didn't, but he should not have).
That wound up being a bad strategy, however; Harden made 87.9% of his free throws during the 2018-19 regular season, which was the 12th best percentage in the league among all 128 qualified players.
Meanwhile, apply this fouling strategy to a 63.7% free throw shooter like Rudy Gobert, or LeBron James — yes, he only made 66.5% of his free throws and was among the worst in the league — and teams start to see better results.
In terms of players in the NBA Finals, Kahwi stands as the lone target (Kevin Durant is still out for Game One) with at least one reviewable foul being called for him a game.
Besides the fact that LeBron is worse from the free throw line compared to Harden, we can conclude that Harden is the focus of these late game situations, as the ball is mostly in his hands. But, what he doesn't admit is that he also causes the most fouls that don't get called.
Among all players in the Last Two Minute Reports post-All Star Break, James Harden is third overall for most incorrect non-calls as a committing players. In other words, in the last two minutes of the nine games we looked at, he committed four fouls that should have been called. That's double the number of fouls that NBA Finalist Draymond Green, who is notorious for being a defensive enforcer, committed during this part of the season.
This breaks from a usual narrative that James Harden has: the flopper. ESPN wrote one feature on how he could draw fouls like an Oscar nominee, with several of the league's players agreeing with his expertise. Yet even here, he still has his shining moments as a bad flopper; on average, he was the disadvantaged player (player who got fouled) 2.22 times a game during correct non-calls. In other words, at least two times a game, Harden tried to draw a foul and didn't get it, or the league looked at personal contact made between him and another player, but concluded it was rightly not called a foul.
As Harden continues to be king of the drawn fouls, a player actually in the NBA Finals — Kahwi Leonard — is among those who gets an NBA Razzie Award for Flopping... Kind of.
In the eight Raptors games listed in the Last Two Minute Reports, Kahwi was the disadvantaged player for 24 total non-calls. That's three a game, which is one of the worst in the league, but considering he also gets the most calls for him in general, this basically means that the Raptors trust him to have the ball in his hands during the last two minutes of a close game. And for good reason: he made 85.4% of free throws in the regular season, and made 133 out of 152 free throws in the Playoffs up until the NBA Finals.
And still, his correct non-call rate is not as bad as eventual free agent Kyrie Irving, who had 26 correct non-calls in seven close games during the end of the regular season.
Combined with two abysmal shooting performances in Games 3 and 4 of the Easter Conference semifinals, one could argue that something is up with 2016 NBA Champion Uncle Drew's ability to draw fouls and make shots. And, in considering how he left for the locker room before the final buzzer in Game 4, a critic might just draw this conclusion outright.
Conclusion: The Games
The Playoffs are, of course, the Playoffs. Harden, Kyrie, and everyone else get a fresh start despite the calls they did (or didn't get) since October, and that regular season record doesn't really matter anymore. What matters are those Playoff calls and non-calls, right?
Or does it?
This question puts us into a rabbit hole here. We could go back and replay every one of those regular season games with every conceivable correct foul call, which may alter the course of the Playoffs itself by drastically changing the seeding of teams, should there be enough change in the results.
Let's start with the ones we've already looked at:
The 76ers and Bulls game that Courtney Kirkland called had that one incorrect call. If Robin Lopez never gets to the line due to that blown call, never hits that one free throw, and all other points remain constant, the game goes into overtime and the Bulls may not have won. That means the 76ers could have ended the season with 53 wins instead of 52... And may have gone into their game against the Rockets two nights later with more momentum.
This is why these Last Two Minute Reports are important; one wrong call can cause a butterfly effect that changes the course of NBA history. But of course, this is only looking at the last part of this year's regular season. What if we looked at all the games for this regular season? What about seasons past, all the way to the first year that the NBA did these reports?
This is where you, the reader and potentially groundbreaking NBA database creator, come into play.
We are well aware that our data isn't complete; we only counted the games following the All-Star break heading into the Playoffs. The Playoffs themselves? We're getting the data for it as you're reading this.
But what about the regular season pre-All Star break? Well, that's where we're going to pull down our curtain and ask for help. We admit that we didn't figure out a way to get this data in a speedy and efficient manner with no errors. The way that the NBA publicly released this data pre-All Star break was fundamentally different, as you can see by clicking around on the Last Two Minute Report website yourself (it was done in a PDF, to be specific).
This leaves us, the community, with two options: either someone with way too much time on her or his hands manually does it, which leaves human error, or someone cracks the code to converting this data efficiently.
So, who can we pass the ball to? We don't know yet, but we know someone in the KG Base community can slam dunk this. And in doing so, they will help us — the community of NBA fans — finally discuss what would have been in this league filled with... Well... Human beings.
Edited by Joshua Fruhlinger and Jon Marino. Data collected by the Thinknum team. Graphs and cover art by John Lee. Gifs by Olga Chernenok. Anthony Tapias contributed in analyzing data. Both LeBron lover Gregory Ugwi and unfortunate Knicks fan Justin Zhen served as inspiration for this piece.